Cheapest 12-second 9:16 clip for a TikTok ad on Sume, by model
TikTok non-Spark ads need at least 540x960 for 9:16. At 720p or 768p a 12-second clip costs $0.90 on MiniMax H3, $1.50 on Wan 3.0 and Omni, $4.54 on Seedance 2.

The cheapest Sume model that makes a 12-second 9:16 clip at a TikTok-safe size is MiniMax H3 at its 768p tier, $0.90, followed by Wan 3.0 and Gemini Omni Flash at 720p, $1.50 each. TikTok's ad specification for non-Spark ads sets a 540x960 minimum for 9:16, so the question is which tier of each model clears it.
The platform minimum
TikTok's video ad specifications page lists, for non-Spark ads: 9:16 at a minimum of 540x960, 16:9 at 960x540 and 1:1 at 640x640; files of .mp4, .mov, .mpeg, .3gp or .avi; up to 10 minutes; at most 500 MB; and a bitrate of at least 516 kbps. If Sume's tier names (480p, 720p, 768p) refer to the short edge, as such names usually do, a 480p vertical clip is 480 pixels wide, under the 540 minimum. Confirm the pixel size of one output before you buy a batch.
| Model | Tier | Per second | 12 seconds |
|---|---|---|---|
| minimax-h3 | 768p | $0.075 | $0.90 |
| wan-3.0 | 720p | $0.125 | $1.50 |
| gemini-omni-flash-1.1 (3 to 10 s only) | 720p | $0.125 | 12 s not valid; 10 s = $1.25 |
| seedance-2 | 720p | $0.378 | $4.536 |
| seedance-2.5 | 720p | $0.5778 | $6.9336 |
A limit that changes the plan
Gemini Omni Flash on Sume accepts 3 to 10 seconds, so it cannot make a single 12-second clip. If you want Omni's native audio, make a 10-second clip, or two clips joined in a Timeline job at $0.10 per output minute. MiniMax H3 accepts 5 to 15 seconds, Wan 3.0 accepts 2 to 30, Seedance 2.0 accepts 4 to 15.
Resolution above the minimum costs more without helping compliance. Wan 3.0 at 1080p is $3.00 for the same 12 seconds, $1.50 more than 720p.
How to run it
Call POST /v1/videos with the model pinned, aspect_ratio set to 9:16 and resolution set to the tier above, and poll the job. Send an Idempotency-Key so a retry does not bill twice. Details are on the Video generation page. Re-read TikTok's specification page before upload, as ad specs are revised.
The platform's rules and the model's output are separate checks. Price tells you what a clip costs; only opening the file tells you it passes.
- Cheapest passing tier: MiniMax H3 768p.
- Omni cannot do 12 seconds in one clip.
- Check pixel size on one sample before the batch.
Reproduce the numbers
Every Sume price on this page comes from the public catalog at https://api.sume.com/v1/catalog, which needs no API key. Fetch it, find the row for the job or model named in the table, and multiply by your own seconds, characters or image count. Sume bills the list price shown in that row, so the catalog figure is the one that applies to your balance.
Vendor figures were read from TikTok Ads: video ad specifications on 2026-10-09 and are the vendor's to change. Where a vendor page did not show a number, this page leaves it out instead of guessing, and any assumption that is ours is labeled as one.
Sources
Related posts
More in Comparisons
- 12 s of Kling 3.0 Motion Control vs plain Wan 3.0 or Seedance 2.5
Kling 3.0 Motion Control is $0.1575 a second on Sume: $1.89 for 12 seconds, against $1.50 for Wan 3.0 at 720p and $6.93 for Seedance 2.5.
- 12,000 Recraft V4 images: fal list $480 vs Sume $600, the $120 gap
Recraft V4 is $0.04 per image on fal and $0.05 on Sume. For 12,000 images that is $480 versus $600. What the 25% covers, and when to call fal directly.
- 15-second clip: Seedance 2.0 vs 2.5 at 720p and 1080p on Sume
Seedance 2.0 tops out at 15 seconds on Sume. A 15-second clip is $5.67 at 720p versus $8.67 on 2.5, and $12.76 versus $21.32 at 1080p.
- 16-megapixel image: Sume upscale $0.20 vs FLUX.3 4k generation $0.607
Sume bills image upscale at $0.20 per image (about 16 MP reserved). BFL lists a FLUX.3 4k image, about 16 megapixels, at $0.607. 100 images: $20 vs $60.70.
Written by Sume